Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    specific actions and observations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    يز
    0.43
    าร์
    0.41
    г
    0.41
     язы
    0.40
    اني
    0.39
    tube
    0.39
     கண்
    0.39
    يل
    0.38
    кие
    0.38
    デー
    0.38
    POSITIVE LOGITS
    并且
    0.51
     marchandises
    0.43
     ideals
    0.42
     Motley
    0.41
     Essence
    0.40
    ocos
    0.40
    并在
    0.39
     Kingdoms
    0.39
     součást
    0.39
    acency
    0.38
    Act Density 0.009%

    No Known Activations